DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://nagoya.apache.org/bugzilla/show_bug.cgi?id=8657>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=8657

java.lang.NullPointerException: XRTreeFrag.str(XRTreeFrag.java:280)





------- Additional Comments From [EMAIL PROTECTED]  2002-04-30 19:47 -------
I agree we shouldn't be throwing the exception. What I'm questioning is just 
whether this is the place where we should be preventing the issue from arising, 
rather than (a) keeping m_dtm.getStringValue(m_dtmRoot) from ever returning 
null, or (b) not letting XRTreeFrag get into a state where that call might 
return null.

I may be mistaken, but I believe (b) is the actual cause of the symptom you're 
trying to address, since your description of an intermittant error appears to 
match the timing dependency which arose between XRTreeFrag's finalizer and 
main-line Xalan execution. The proposed patch will mask the symptom. But the 
symptom is diagnostically useful for tracking and curing the actual disease, 
which also has other undesriable effects and should be addressed at that level.

It's a good workaround patch for the symptom. But -- again, unless I'm 
misreading the situation -- it would have adverse effects on our ability to 
actually debug the system.

Reply via email to